[00:36] ok, I've managed to get a static quota set... but the goal is to have user quotas set through mysql... [00:37] my mysql sucks too, the query is returning the correct value to overide the static quota but dovecot doesn't apply it... [00:43] internalkernel: wouldn't that be dovecot's fault, then? [00:43] that's the line I'm following... has to be a config problem... [00:43] more correctly, my fault.. :) [00:44] I don't think I've told dovecot where to find the user override values... and this whole thing is not well documented... [00:45] Im learning by breaking... [00:51] Hallelujah... [00:52] internalkernel: you should patch the docs [00:53] I was setting maildir:storage="value" is quota - when it should have been *:bytes="value" is quota_rule [00:53] docs? they have those? [00:53] :) [00:53] I've been documenting... so a proper wiki is in the works... [01:09] internalkernel: the docs are linked from /topic [02:51] New bug: #512167 in apache2 (main) "apache overriding php's error_log setting" [Undecided,New] https://launchpad.net/bugs/512167 [03:07] Is it possible to move more than one folder at a time with rsync [03:09] I don't think so, I think it's like cp in that sense - you can start multiple instances but only source => destination [03:27] I got it [03:27] internalkernel: I wanted to go mulitple folders backed up to a single [03:27] rsync source1 source2 source3 ... target [03:29] Xpistos: is this local? [03:29] I don't think you can do multiple sources, but you could test it... [04:06] New bug: #512110 in krb5 (main) "gssd regression, "Program lacks support for encryption type"" [Undecided,New] https://launchpad.net/bugs/512110 [05:28] Grr [05:29] Doesn't anyone use (and document) how to manage cups without resorting to some goddamn web UI? [06:18] twb: I am unaware of a way to manage cups without using the web UI.. but the web UI has worked alright for me. I'm not sure if they offer another way to manage it other than the web UI. [06:31] jasonb: all the web UI does is write to files in /etc/cups [06:31] But rather than making a snapshot, using the web UI, then diffing, I'd rather just bloody well know how to edit the files directly [06:40] twb: Yeah, it's possible it works that way. I haven't seen any tutorials about editing files in /etc/cups. [06:44] Thanks anyway === erichammond1 is now known as erichammond [07:11] I'm having problems troubleshooting a hard locking situation. My server will hard lock after about 10 minutes of being up after a reboot. I've looked through logs and I can't find anything indicative of a problem. I also had the datacenter swap out the hardware because I assumed it was a hardware issue, but it still happens [07:13] Also, according the datacenter staff, once it was hardlocked there was no video display and the keyboard and mouse would not respond at all [07:13] Er, no mouse :p [07:15] It actually looks as if it's almost exactly 10 minutes after booting that it will hard lock === deegee_1 is now known as deegee [08:44] good morning [09:03] jumbers: is the klog still available after you reboot? [09:04] jiboumans: if you're still using HDDs, run smartctl on them? (That's a long shot, though.) [09:04] twb: i guess that was meant for someone else [09:04] Sorry, yes. Bad completion. [09:05] twb: I think I fixed it [09:05] jumbers: what was it? [09:05] I tried this and it worked: http://serverfault.com/questions/80520/ubuntu-server-9-10-freezes-up-after-10-minutes/91344#91344 === baffle_ is now known as baffle [09:09] twb: ^ [09:15] How do I have my server recognized over my lan as its hostname rather than its IP address? [09:16] entrooo: you have 2 options: 1) usual DNS 2) zeroconf/bonjour/rendezvous (aka multicast-dns) [09:18] do I need a separate DNS server or can I try and setup up bind on my server to broadcast itself? [09:24] ttx: ping? [09:25] jiboumans: pong [09:25] hey, can i give you a quick call? i have a few UEC questions from sg i'd like to answer [09:25] jiboumans: sure, 5min, I'll fetch a coffee first [09:26] ttx: sure, ping when ready [09:26] anyone here use gdisk (gpt fdisk)? [09:29] jiboumans: shoot [09:29] * jiboumans aims [09:56] !anyone > martin- [09:56] martin-, please see my private message [09:57] I always get mdns and whatever avahi provides mixed up. [09:57] Hmm, apparently avahi *is* mdns [09:58] The other thing is the one that gives a route for a useless private address range on IPv4. [10:11] is it possible to resize gpt partitions with gdisk (gpt fdisk)? [10:17] martin-: why don't you use parted? [10:18] And partitions generally aren't resized ever -- they're simply deleted and re-created [10:18] parted won't resize the partition because it has a ext4 filesystem [10:19] That's parted being an arse. [10:19] I thought it was possible to let parted resize the actual partition, and use resize2fs for the filesystem, but parted won't let me do that [10:19] probably [10:19] martin-: yes, that's what I mean by "being an arse" [10:19] yep [10:19] that's why I'm trying gdisk [10:19] Sorry, I misread. I meant parted's idea of a resize is to resize BOTH the filesystem and rewrite the partition table. [10:20] What you should do is (if shrinking) resize2fs -M, then (always) use parted to delete the partition and create a new one with the same start point but a different endpoint, then (if growing) resize2fs. [10:20] Oh, and next time, bloody well use LVM [10:21] so I can delete and recreate the same partition at the same start point, just like good old fdisk? [10:21] and without loosing data? [10:22] If the partition isn't in use, sure. [10:22] Well, let me qualify that [10:22] I haven't actually TRIED that on gpt, but I'm not aware of any reason it should fail. [10:23] Of course, since you've been taking backups every day since you were given the privilege of your very own workstation, and you verify their accuracy with a test restore every month, it will be no problem at all even if you blat the entire disk. [10:23] hehe, of course [10:24] so I can't blame you or your cat? :P [10:24] You can blame whoever you damn well please. [10:25] To be honest, I'd be inclined to blow away the disk anyway and restore it, using LVM this time. [10:29] twb: your suggestion seemed to work, thank you [10:29] managed to recreate the partition without destroying the data [10:29] oh, and yes, I will use LVM next time :-) [10:35] that being said, I hope the parted developers decide to remove the filesystem functionality in the future [10:36] blending filesystem and partition functions kind of breaks the UNIX philosophy of one tool per job [10:51] New bug: #512271 in krb5 (main) "kinit crash" [Undecided,New] https://launchpad.net/bugs/512271 [13:05] mornig [13:05] morning [13:36] New bug: #512317 in mysql-dfsg-5.1 (main) "Database crashes when running "create as select" with no schema specified" [Undecided,New] https://launchpad.net/bugs/512317 [13:44] mornings [13:53] ttx: it looks like we might have problems with virtio again in lucid [13:54] bug #511620 [13:54] Launchpad bug 511620 in qemu-kvm "write errors on virtual disc during install" [Undecided,New] https://launchpad.net/bugs/511620 [13:54] that would be the recently-updated qemu-kvm [13:56] about to start Deploying Ubuntu Enterprise Cloud training [13:56] aubre: sounds cool :) [13:57] yeah, I'm pumped :) [14:26] New bug: #510587 in bind9 (main) "Bind/named does not initialize on boot due to missing IPv6 address" [Low,Incomplete] https://launchpad.net/bugs/510587 [14:26] New bug: #512209 in clamav (main) "clamscan crashes if started with umask u=rw,go= on x64" [Low,Confirmed] https://launchpad.net/bugs/512209 [14:42] Hi! Is there any tool to manage a Postfix queue? [14:48] acalvo: postqueue [14:48] Mostly if you think you need to manage a postfix queue, you probably need to reconsider. It's usually not needed. [14:48] There is also postsuper [14:50] mdeslaur, jdstrand: Have any of you succesfully run the fetchmail test thing from q-r-t recently? [14:51] soren: not recently [14:51] soren: is it busted on lucid? [14:52] jdstrand: I'm running it on Karmic at the moment. It fails. [14:52] soren: last update we did to fetchmail was aug 2009, that's probably the last time it was run/tested [14:52] hmmm... [14:52] jdstrand: It tries to access dovecot's auth socket, but it's not configured. [14:52] let me try [14:54] thank you ScottK, it seems what i was looking for [14:55] soren: did you install 'netbase dovecot-imapd dovecot-pop3d python-openssl ssl-cert'? I don't see those as listed in QRT-Packages, which may be why it is failing [14:56] New bug: #375490 in spamassassin (main) "Spamassassin fails to escape backslashes for latest postgres input syntax" [Undecided,Incomplete] https://launchpad.net/bugs/375490 [14:57] jdstrand: I did not have python-openssl, but installing it didn't help. [14:58] jdstrand: I think I understand the failure. I do not completely understand how I'm supposed to fix it. I've never touched these scripts before. [14:58] jdstrand: It fails because it calls into testlib_dovecot, which spends a lot of time trying to access dovecot's auth_socket, but dovecot is not configured to have an auth socket by default, so naturally, this fails miserably. [14:59] soren: hrmm.. kees added those tests, maybe he didn't commit the changes to testlib_dovecot.py? [14:59] Conceivably. [15:00] soren: we could spend a lot of time trying to figure it out, or ask kees when he comes online, since it apparently worked for him at some point [15:00] I opt for the latter [15:00] jdstrand: /me too :) [15:01] :) [15:01] jdstrand: I expects kees' name has been mentioned enough in this conversation to pique his interest. [15:01] soren: probably so. kees is generally pretty good at reading backscroll [15:01] * jdstrand waves to kees :) [15:02] (had to get two more references to kees in :P ) [15:02] jdstrand: Yeah, you can't mention kees too often. :) [15:02] hehe [15:03] * soren sort of calls it a day (been at it since 6 AM), but will be back to see what kees has to say about the those things of kees' that we talked about we would get kees' input on [15:03] * soren wanders off [15:21] New bug: #498632 in clamav (main) "package clamav-freshclam 0.95.3+dfsg-1ubuntu0.09.10 failed to install/upgrade: problém se závislostmi - nechávám nezkonfigurované" [Undecided,Incomplete] https://launchpad.net/bugs/498632 [15:52] smoser: around ? [15:53] here [15:53] smoser: just to let you know, I played with the boothooks stuff thios morning [15:53] smoser: mostly worked :) [15:53] well, thats better than "fell all over the floor" [15:54] i'm very interested in feedback, and i realize that the first bit is "examples" [15:54] :) [15:54] smoser: right. Looking into the code to know how to write the stuff isn't exactly lowering the barrier of entry [15:54] smoser: only issue I found is that there is a window where the hostkeys aren't replaced yet [15:55] really? [15:55] smoser: On UEC I got a hostkey warning [15:55] yes, I know the IP in advance, so I can ssh as soon as it says "running" [15:56] so the reason for that is that i'm not starting early enough, and not blocking ssh [15:56] usually will pick up ssh as soon as it's run [15:56] so the normal ssh starts up [15:56] yes. [15:56] then keys are changed and ssh restarted [15:56] i can easily work around that by removing the keys in the build process [15:56] then the first ssh will fail [15:57] smoser: that would be better, yes [15:57] i hadn't seen the issue, but knew it was possible. [15:57] smoser: want a bug ? [15:57] sure. [15:57] against ec2-init, I suppose. [15:57] ther is no better place for the ec2-automated-builds where it will actually go in [15:57] smoser: didn't try "include" yet, was missing some syntax and got lazy [15:58] so for an include file: [15:58] #include [15:58] http://some.url [15:58] http://some.other.url [15:58] one per line. simple. [16:03] smoser: bug 512377 [16:03] Launchpad bug 512377 in ec2-init "ssh is started before hostkey is installed/regenerated" [Undecided,New] https://launchpad.net/bugs/512377 [16:52] grr nickserv === roobiew_ is now known as robbiew [17:24] I feel like you guys were maybe talking about me. [17:25] jdstrand, soren: ^^ :) [17:25] heh === erichammond1 is now known as erichammond === rberger_ is now known as rberger [18:14] mathiaz: ping? [18:14] jiboumans: pong [18:14] jiboumans: but this game is out-dated dude! [18:14] mathiaz: best game ever [18:15] mathiaz: so, pdx sprint. we have some puppet stuff on the agenda [18:15] mathiaz: pdx is also the hometown of the puppet guys; how useful would be some facetime, how useful would be some hacking time? [18:16] jiboumans: right - I don't see any puppet item on the Server agenda though. But some face time with Luke would be useful [18:16] jiboumans: I've got some ideas to ask him [18:16] jiboumans: I've got some ideas to *discuss* with him [18:17] mathiaz: the agenda is what we make of it [18:17] in fact, i think ttx sent around a mail askign for input :) [18:17] jiboumans: which are actually related to the puppet specs [18:17] mathiaz: so i can get luke for a beer (already asked) but he's out of town most of it. we can get some of the minions though if i ask nicely [18:17] jiboumans: right - I've added some things already - I plan to add more :) [18:18] jiboumans: hm - a beer with luke should be enough to discuss what I have in mind [18:18] ok; monday or tuesday evening then [18:18] jiboumans: works for me [18:18] mathiaz: awesome [18:19] jiboumans: there is one specific feature we've asked for [18:19] jiboumans: that should be part of 0.25.4 [18:20] jiboumans: so depending on how integrating 0.25.4 moves forward it may be worth having access to some puppet devs [18:20] .. but? [18:20] alright [18:20] jiboumans: this is part of the puppet-etckeeper integration spec [18:20] jiboumans: and soren is currently assigned for that [18:20] the root-thing? [18:20] jiboumans: hm - the hooks integration [18:21] jiboumans: making sure that the puppet client calls etckeeper before and after its runs [18:21] jiboumans: upstream puppet should have provided the hooks to do that [18:21] jiboumans: we need to right the actual calls to etckeeper [18:21] ok, is that actually a topic that'd need discussing with them? [18:22] jiboumans: depending on how this works out, it may be useful to have access to some puppet devs [18:22] jiboumans: at first glance I don't think so [18:22] jiboumans: having someone "on-call" may be useful ;) [18:23] alright [18:23] jiboumans: if we run into some issues [18:23] do we know if the hooks have landed already? [18:23] jiboumans: they should be part of 0.25.4 which is what upstream would like to be included in LTS [18:24] jiboumans: http://projects.reductivelabs.com/issues/2914 [18:24] ie'fixed' [18:24] jiboumans: it should be part of 0.25.4 [18:44] How can I connect to my ubuntu server with its hostname rather than its IP on my LAN without having a separate DNS server? [18:44] entrooo: install avahi, I think [18:45] thank you, I will look into it [19:12] zul: hi! [19:13] mathiaz: hey how goes it? [19:13] zul: is the mysql upstart job available somewhere? [19:13] zul: (the one supposed to replace mysqld_safe) [19:13] mathiaz: yep its in lucid right now [19:14] zul: great - thanks [19:14] mathiaz: np === Barre_ is now known as Barre [19:44] kees: Maybe. :) [19:45] kees: Any idea what's going on there (with the fetchmail test)? [19:45] mathiaz, possibly soren, we have a date for tuesday afternoon going into the evening [19:45] A... date? [19:45] soren: yeah, wear something pretty, i'll pick you up at 8 [19:45] soren: with a puppet! [19:46] * soren looks around for candid cameras [19:46] soren: 'strue i'm afraid [19:47] soren: i asked luke kanies to join us at the sprint in pdx for some puppet related stuff [19:47] Oh, *next* Tuesday. [19:47] he + 1 or 2 will drop by tuesday afternoon, then we have some beers [19:47] * soren is up for beer [19:47] soren: yeah right, like *that* would have explained everything :) [19:47] I have a presentation Wednesday morning, though.. [19:47] Meh, I'll wing it. [19:47] soren: i dont expect this to be an all night drinky thing [19:47] Aw. [19:47] soren: basically this is pay geeks in beer ;) [19:48] well, you *can*, but that's not what i'm *making* you do :) [19:48] That's not what my activity report will say, but ok. [19:48] soren: i expect your activity report will be reproducable by closely connecting keyboard + face repeatedly [19:50] Not so much face as horns on our helmets. [19:50] Close, though. [19:50] vikings++ [19:51] * RoyK throws a snowball in soren's direction [19:52] Can I use evolution-server with Outlook? [19:52] For sharing contacts and calendar? [19:52] Or MUST everyone switch to Evolution at the same time? [19:52] Er.. [19:53] Can I have an environment with both Outlook and Evolution users using the same evolution server? [19:53] I think you mean to ask this in #ubuntu. [19:53] There's no such thing as evolution-server [19:53] soren, evolution-data-server - evolution database backend server [19:54] There's evolution-data-server, which is just a service evolution uses as it's backend. It's not shared among users. [19:54] Does evolution have anything to sync contacts via Google? [19:54] Ok. [19:54] ...and it's not really on-topic in here. [19:54] Sorry [19:54] ScottK: :) [19:54] Ok. So I probably need to set up an OpenLDAP server for this instead, right? Can I get help with that here then? [19:55] tdn: It's on-topic, certainly. [19:55] Whether anyone can/will answer your questions: Only time will tell. [19:56] I have been trying to set up an OpenLDAP server a few times. I have read some guides/howtos, but either I just can't wrap my mind around how it works, or I am doing something wrong, because I cannot get it to work. I thought it might be fairly simple to set up an LDAP server for contacts. [19:57] * RoyK informs the world that this is off topic, but far too seriously hilarious to be left out http://www.guardian.co.uk/books/2010/jan/25/oral-sex-dictionary-ban-us-schools?DSF [19:58] Where do I start? Will someone guide me a little? [19:58] RoyK: ara you certain you want to post an external url - *these days* ? [19:58] *are [19:59] most surprising is that it's *not* the onion [19:59] guntbert: huh? [20:00] RoyK: we had a lot of spam on freenode lately - mostly from users who couldn't resist clicking on a link in IRC - and getting klined promptly [20:00] hehe [20:00] seen it [20:01] mathiaz: is you udw thing soon? [20:01] zul: yes - now [20:01] I guess maybe, perhaps, something, people might find guardian.co.uk familiar [20:02] mathiaz: wohoo...let the party begin [20:04] RoyK: I'll never know :) [20:06] smoser: virt config patch has been applied to Lucid it will show up in the next kernel upload [20:06] good deal, thanks jjohansen [20:45] hi, how to check preseed file for openoffice or something like this after system install ? [20:47] mealstrom: on server? would surprise me wildly [20:50] guntbert: change package openoffce for vsftpd or sshd, (I've got about 20 servers and 160 desktops) and want some basic autodeployment. [20:51] mealstrom: ok sorry - no autodeployment experience here :) === cyphermo1 is now known as cyphermox [21:05] I've been researching avahi, but cannot figure out how to make my server be recognized on my LAN by its hostname rather than just its IP address, any suggestions of what software I should use? [21:14] mathiaz: who from our team talks to the bacula folks the most? [21:28] just installed lucid alpha2 server in a kvm, but it doesn't want to start up. any hints? [21:32] after restarting from install, just two lines appear in the vnc, fsck from util-linux-ng, then /dev/sda1 clean, and that's it [21:47] hm, never mind. seems like it doesn't clean up after itself. it booted ok, but I didn't get the login screen [21:48] only after going to alt+f2 I noticed [21:56] jcastro: hm - zul [21:57] cool [21:57] jcastro: upstream attended UDS jaunty as well [21:57] jcastro: so I meet him there [21:57] ok [21:57] zul: I've pinged him about application indicators [21:57] which as it turns out has nothing to do with the server bits. [21:58] however it would be cool if the little tray thing was using the app indicators for lucid, so I've asked him if he needed anything [21:58] unsurprisingly no one was screaming with joy over a panel icon for a server product. :p === erichammond1 is now known as erichammond === erichammond1 is now known as erichammond === erichammond1 is now known as erichammond [23:02] i was looking at the 'Canonical' ubuntu 9.10 for ec2. Do they execute the data in UserData on startup like the older alestic images (?) [23:15] smoser: ^ [23:41] hi, anyone can suggest an opensource ticketing system.. [23:41] for support or anything request on the company.